Assessing the impacts of marine protected areas on wrasse populations in Norway MOLLY REAMON SUPERVISORS Lars Korslund Kim Halvorsen University of Agder, 2020 Faculty of Engineering and Science Department of Natural Sciences Abstract Several species of wrasse (Labridae) are increasingly harvested to be used as cleaner fish in salmon aquaculture. The intensity of harvesting wrasse has raised concern regarding sustainability and there is a lack of knowledge on the ecological impacts of such harvesting in Norway. The establishment of marine protected areas (MPAs) could be an effective management tool for maintaining wild wrasse populations. This study investigated the impacts of MPAs on wrasse by comparing body size and catch per unit effort (CPUE) between MPAs and nearby fished areas in Austevoll on the west coast and in Tvedestrand on the south coast of Norway. Age of ballan (Labrus bergylta) was also compared between areas in Austevoll. Sampling in Austevoll was conducted during three sampling periods in 2018-2019; and sampling in Tvedestrand was conducted every June from 2010-2019. This study also investigated a possible non-invasive method of aging ballan using scales by comparing scale age to otolith age. In Austevoll, corkwing wrasse (Symphodus melops) were found to be significantly larger in MPAs while CPUE of all wrasse species and age of ballan did not differ between areas. In Tvedestrand, ballan and rock cook (Centrolabrus exoletus) were significantly larger in MPAs, while CPUE of rock cook and goldsinny (Ctenolabrus rupestris) was significantly lower in MPAs. The larger body sizes found in MPAs indicated a positive effect of protection on size for these species and is consistent with the prediction that there should be larger fish inside MPAs. The lower CPUE of smaller wrasse species in the Tvedestrand MPA contradicted the prediction that there should be more individuals inside MPAs and could possibly be explained by increased predation within the MPA. Furthermore, the scales used to age ballan were found to have a high error rate and a tendency to show a younger age than otoliths, however scales were found to be accurate for aging young ballan (< 6 years). The results from Austevoll demonstrated the ability of MPAs to restore harvested populations within one year of MPA implementation, while the results from Tvedestrand highlighted the potential for indirect effects of protection to occur over a longer period of time. The overall findings suggest that MPAs can be used as a management strategy to maintain natural size structure and abundance of harvested wrasse populations. Recommendations for conservation and management are provided including 1) implementing additional MPAs in Western Norway where intensity of harvesting is highest; and 2) revising current management of the fishery to reflect the life history strategies of wrasse species. 2 Table of Contents 1. Introduction………………………………………………………………………………..6 1.1 Background……………………………………………………………………………6 1.2 The wrasses……………………………………………………………………………6 1.3 Norwegian wrasse fishery……………………………………………………………..9 1.4 Selective fisheries and marine protected areas………………………………………10 1.5 Scale and otolith aging……………………………………………………………….12 1.6 Study objective……………………………………………………………………….13 2. Materials and methods………………………………………………...…………………15 2.1 Study area………………………………………………………….…………………15 2.2 Sampling………………………………………………………….………………….17 2.3 Age determination………………………………………………...………………….19 2.3.1 Flødevigen otoliths and scales……………………………………………...19 2.3.2 Austevoll scales…………………………………………….........................21 2.4 Data analysis……………………………………………............................................22 3. Results……………………………………………………………………………………25 3.1 Scale aging methodology…………………………………………………………….25 3.2 Austevoll……………………………………………………………………………..26 3.2.1 Age…………………………………………………………………………26 3.2.2 Body size…………………………………………………………………...28 3.2.3 CPUE……………………………………………………………………….31 3.3 Tvedestrand…………………………………………………………………………..33 3.3.1 Body size…………………………………………………………………...33 3.3.2 CPUE……………………………………………………………………….36 4. Discussion………………………………………………………………………………..40 4.1 MPA effects on wrasse………………………………………………………………40 4.1.1 Austevoll…………………………………………………………………...40 4.1.2 Tvedestrand………………………………………………………………...42 4.2 Wrasse populations in Southern and Western Norway………………………………43 4.2.1 Limitations of comparing regions………………………………………….43 4.2.2 Geographical variations for the different wrasse species…………………..45 3 4.3 Scale aging methodology…………………………………………………………….46 4.4 Management implications……………………………………………………………47 5. Conclusions and future directions………………………………………………………..51 References……………………………………………………………………………………52 Appendices…………………………………………………………………………………...61 Appendix A: Reported landings of wrasse in Southern and Western Norway…………..61 Appendix B: Maps of sampling locations………………………………………………..62 Appendix C: Accuracy of scale readings………………………………………………...64 Appendix D: Length distribution plots…………………………………………………..65 Appendix E: Model selection tables……………………………………………………..69 Appendix F: Methods for generalized additive models………………………………….72 4 Preface I would like to thank my supervisors, Kim Halvorsen and Lars Korslund, for providing invaluable guidance and checking in on me throughout the entire process. Meetings with Kim and Lars were full of enthusiasm and I always left feeling motivated and excited about the next step. Your extensive knowledge and experience with statistical analysis, modelling, and academic writing has been an enormous help to me. Thank you for always being accessible for advice, for reassuring me that no question is too big or too small, for providing feedback on drafts, and for always encouraging me. I am grateful to Torkel Larsen for more reasons than I can list on one page. Thank you for driving the boat during fieldwork, for instructing me how to properly throw and haul a fyke net, and for teaching me everything I know about identifying wrasse and other fish species. Thank you for remaining calm whenever I misidentified corkwing as goldsinny and for having the patience to teach me how to tie the boat knot even though I failed countless times before I succeeded. Thank you for showing me how to prepare scales in the lab and for instructing me how to read the age of scales and otoliths. The help and support you have provided me during this year has been instrumental towards this thesis. Thanks also to Anne-Berit Skiftesvik for her assistance with species identification during my first round of fieldwork and to Ben Ellis for hauling the fyke nets during my second round of fieldwork. Thank you to Sigurd Espeland for sharing the MPA data from Tvedestrand with me and for providing information about the survey. Lastly, I would like to thank my friend Marthe Ruud for embarking on this wrasse journey with me. From keeping me company during lab work to keeping me calm when trying to navigate statistics in R, I have appreciated having you alongside me every step of the way. Kristiansand 28.05.20 Molly Reamon 5 1. Introduction 1.1 Background Several species of wrasse (Labridae) are increasingly fished in Norway and the British Isles to be used as cleaner fish in the farming of Atlantic salmon (Salmo salar Linnaeus, 1758) and rainbow trout (Oncorhynchus mykiss Walbaum, 1972) (Halvorsen et al., 2017; Skiftesvik, Durif, Bjelland & Browman, 2015). The wrasse fishery in Norway began in the late 1980s after it was discovered that wrasse may be used in salmon farming to remove sea lice (Lepeophteheirus salmonis Krøyer, 1837) (Bjordal, 1988; Skiftesvik et al., 2014). The demand for wild wrasse in Norway has increased substantially since 2009, after the salmon lice began to develop resistance to the pesticides most commonly used to treat salmon lice (Skiftesvik et al., 2014; Nilsen, 2008). Norway is the world’s leading producer of salmon (FAO, 2019) and outbreaks of lice are among the greatest challenges facing the industry (Svåsand et al., 2017) suggesting that the wrasse fishery will continue to be relevant in the near future. The intensity of harvesting wild wrasse has raised concern among the scientific community regarding sustainability and there is a general lack of knowledge on the ecological effects of such harvesting in Norway (Espeland et al., 2010; Skiftesvik et al., 2015; Halvorsen et al., 2017; Olsen, Halvorsen, Larsen & Kuparinen, 2018). 1.2 The wrasses Wrasse are a family of marine fish (Labridae) of more than 500 species that are numerous among inshore areas (Skiftesvik et al., 2015). There are five common species of wrasse found in Norwegian waters including corkwing (Symphodus melops Linnaus, 1758), goldsinny (Ctenolabrus rupestris Linnaeus, 1758), ballan (Labrus bergylta Ascanius, 1767), cuckoo (Labrus mixtus Linnaeus, 1758), and rock cook (Centrolabrus exoletus Linnaeus, 1758) (Costello 1991; Figure 1). The scale-rayed wrasse (Acantholabrus palloni Risso, 1810) also exists in Norwegian waters however it is seldom observed because it inhabits deep waters (50- 270 m) (Costello, 1991). Wrasses are intermediate predators that feed on mollusks and crustaceans (Costello, 1991) and are prey for larger predators such as gadoids and seabirds (Nedreaas et al., 2008; Steven, 1933). The wrasses exhibit a variety of life history strategies which will be reviewed in the following paragraphs. 6 Figure 1: The six wrasse species found in Norwegian waters including ballan, corkwing (top male, bottom female), rock cook, goldsinny, cuckoo (top female, bottom male) and scale-rayed wrasse. Illustration